prov/efa: fix leak of dmabuf fd in cuda p2p probe #10708
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
prior to this patch, when efa_hmem_info_check_p2p_support_cuda elected to attempt dmabuf for p2p, we previously leaked the file descriptor returned by cuMemGetHandleForAddressRange in all cases. This ultimately meant the dmabuf stuck around for the lifetime of the process, even after dereg and after releasing the memory back to the device mempool. All calls to cuda_get_dmabuf_fd need a corresponding close call. Signed-off-by: Nicholas Sielicki <[email protected]>
shijin-aws
reviewed
Jan 20, 2025
shijin-aws
approved these changes
Jan 20, 2025
|
AWS CI failure is likely an infra issue. Will fix it and retest |
|
bot:aws:retest |
|
In #10716, I have defined ofi_hmem_put_dmabuf_fd() for CUDA which could be used instead of close(). |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
prior to this patch, when efa_hmem_info_check_p2p_support_cuda elected to attempt dmabuf for p2p, we previously leaked the file descriptor returned by cuMemGetHandleForAddressRange in all cases. This ultimately meant the dmabuf stuck around for the lifetime of the process, even after dereg and after releasing the memory back to the device mempool.
All calls to cuda_get_dmabuf_fd need a corresponding close call.